OBJECTIVE: To evaluate the effect of a school based physical education (PE) program and the amount of leisure time sport (LTS) on children's bone health and to examine if LTS influences the impact of school type on children's bone health. METHODS: Children attending "sports" schools (6×45min PE lessons per week) were compared to children at "traditional" schools (2×45min of PE lessons per week) in Svendborg, Denmark. Whole-body DXA scans were performed at baseline (2008) and at a two-year follow-up (2010). Bone mineral content (BMC), bone mineral density (BMD), and bone area (BA) were measured. Multilevel regression analyses examined the impact of school type and LTS participation on bone. RESULTS: 742/800 (93%) invited children accepted to participate. 682/742 (92%) participated at two-year follow-up. Mean (SD) age was 9.5years (0.9) at baseline. A positive association between LTS and BMC, BMD (p